SmartX, a leading provider of modern IT infrastructure products and solutions, has officially released version 1.3 of its production-ready container management and service product, SMTX Kubernetes Service (SKS). Designed for high-performance computing scenarios, the new release introduces support for physical machines as cluster nodes, enabling users to better utilize server resources and increase business agility. Additionally, enhanced visualization features in the user interface make it easier to manage the full lifecycle of Kubernetes clusters and applications. With the updates, SKS 1.3 greatly improves the user experience of container management on SmartX Enterprise Cloud Platform (SmartX ECP).
Release Background
SKS offers critical Kubernetes service for SmartX ECP. Leveraging industry-leading virtualization, distributed storage, and networking security components from SmartX, SKS enables enterprises to easily deploy, manage, and operate production-ready Kubernetes clusters across various environments.
With the advantages of SmartX virtualization, such as resource efficiency, elastic scalability, and secure isolation, SKS offers Kubernetes clusters in virtualized environments for users with the following needs:
- Rapid deployment and flexible scaling of Kubernetes clusters (e.g., development and testing environments).
- Providing dedicated Kubernetes runtime environments for multi-tenant scenarios.
- Supporting both virtualized and containerized applications within limited resources, while ensuring seamless network connectivity between VMs and containers with unified security management.
- Building various types of Kubernetes clusters to balance performance and cost management requirements.
SKS 1.3 offers a more comprehensive container infrastructure solution by enabling unified management of both virtual machines and containers within Kubernetes clusters, whether based on virtual machines or physical machines.
Unified management of VM and container environments
Physical Machine-Based Kubernetes Clusters
Building on its existing support for virtual machine-based Kubernetes clusters, SKS 1.3 introduces the ability to create Kubernetes clusters on physical machines. This feature is especially beneficial for users who prefer deploying Kubernetes clusters directly on physical machines in the following scenarios:
- When business applications demand extreme performance (e.g., AI scenarios), eliminating virtualization overhead can improve the efficiency of computing resources.
- Repurposing older servers to enhance asset utilization and reduce costs.
However, users running physical machine-based Kubernetes clusters often have their Control Plane nodes hosted independently on a vendor’s virtualization platform. This lack of unified infrastructure management poses various challenges. SKS 1.3 addresses these issues by effectively supporting physical machine-based Kubernetes clusters.
Challenges | SKS 1.3 Response Strategy |
Fragmented Operations Management: Virtualized and physical machine resources rely on different vendors and management platforms, leading to slow issue identification and lengthy troubleshooting processes, which in turn reduces operational efficiency. | SKS 1.3 offers a comprehensive infrastructure solution supported by a single vendor, covering virtualization, storage, and networking to meet the needs of both virtual machine and physical machine Kubernetes clusters, enabling unified management. |
Fragmented Usage: Due to the lack of proper configuration and optimization of virtual machines and their distributed storage by container vendors, the performance and stability of Kubernetes clusters are affected:
|
To enhance the reliability of the Kubernetes Control Plane, which is deployed on hyper-converged virtual machines, SKS 1.3 introduces the following features:
|
Key features of physical machine-based Kubernetes clusters supported by SKS 1.3:
- Automatic Control Plane Node Creation: Kubernetes Control Plane nodes are automatically created and run on SmartX hyper-converged virtual machines, eliminating the need for users to preconfigure virtual machines.
- Architecture Support: SKS 1.3 supports x86_64 and AArch64 architecture physical machines as Worker nodes for the workload cluster.
Work node clusters based on virtual machines or physical machines
Users can choose between virtual machine and physical machine clusters accordingly, enhancing flexibility and reducing operational costs. The key benefits include:
- Flexible Compute Resource Selection: Users can select appropriate compute resources based on business requirements. Virtual machine clusters are ideal for scenarios where resource utilization and management flexibility are critical, while physical machine clusters offer more powerful compute capacity, serving as a complement to virtual machine clusters for latency-sensitive applications.
- Simplified Operations: Kubernetes clusters can be built on different underlying hardware architectures and managed centrally, reducing operational complexity and improving management efficiency.
- Improved Server Utilization: By consolidating existing hardware resources, SKS 1.3 allows for unified management of both virtual and physical machine clusters, increasing overall server utilization. For users looking to repurpose existing servers, resources can be allocated flexibly between virtual machines and physical machines according to workload characteristics, maximizing cost efficiency and optimizing resource allocation and cost control.
For detailed information, please refer to the introduction and demo video.
Kubernetes Full Lifecycle Resource Management
In most enterprise IT environments, virtual machines and containers coexist, and mixed workloads are typically managed through separate interfaces. For example, the infrastructure resources required for Kubernetes clusters (such as virtual machines, storage, and networking) are managed through the infrastructure vendor’s interface, while applications running on Kubernetes clusters are managed through third-party container cloud platforms. This fragmented management approach leads to a disconnect between the infrastructure resources and applications of Kubernetes clusters, increasing operational complexity and costs.
SKS 1.3 addresses this issue by enhancing its Kubernetes infrastructure management with the addition of Kubernetes application visualization features. Through a unified graphical interface, users can perform full lifecycle management of Kubernetes resources such as Deployment, Pod, and Persistent Volume, across both virtual machine and physical machine clusters. This eliminates the need to switch between different management platforms, allowing users to configure and manage both virtual machine and container-based applications, along with their associated compute, storage, and network resources, ultimately improving the efficiency of full-stack IT operations.
For detailed information, please refer to the introduction and demo video.
Use Cases
Since its official release, SKS has been deployed in production environments across various industries, helping users build production-ready Kubernetes clusters and achieve unified management of both virtualized and containerized environments.
Zigong First People’s Hospital: The hospital deployed SKS through a one-click installation on its unified cluster management platform. Using a single SmartX management platform, the hospital now manages both its virtualized and containerized workloads, achieving seamless interconnection between virtual machines and containers in a flat network architecture. Additionally, the platform enables unified management of network security policies.
A China State-Owned Water Utility: The utility built an SKS cluster on a hyper-converged infrastructure to support all containerized applications and middleware for key business processes, including residential water supply, wastewater treatment, online billing, and office automation (OA), with over 140 Pods in total. Using the SmartX network plugin, the company achieved seamless connectivity between container and virtual machine networks. SKS simplified day-to-day operations and significantly accelerated Kubernetes cluster deployment.
An Autonomous Driving Company: After adopting SmartX’s full-stack hyper-converged solution, the company streamlined the cluster lifecycle management process with automation and gained the ability to flexibly expand or reduce resources based on business needs. SKS enabled unified management of mixed container and virtual machine deployments, improving resource utilization and operational efficiency.
More Information
Download SmartX ECP Product Brief Flyer
Watch Kubernetes AI Use Case Demo Video
Watch Kubernetes Cluster Management Demo Video
Read Blog: Introducing SmartX VCCI Solution: Efficient Integration of VM and Container Resources